Uptime-Kuma Status Extractor

Point at any self-hosted Uptime-Kuma status page and export its monitors, latest status, ping and 24-hour uptime as clean, structured rows.

Uptime-Kuma is the dominant self-hosted uptime monitor. Every public Uptime-Kuma status page — regardless of version or theme — exposes the same unauthenticated public JSON API:

• GET /api/status-page/{slug} — the page config + monitor tree (publicGroupList).
• GET /api/status-page/heartbeat/{slug} — per-monitor heartbeat history + 24h uptime.

This actor speaks that one standard, so a single actor works against every Uptime-Kuma instance in existence — no per-host scraper.

Give it a status page URL; get back a flat, fully-typed dataset row for every monitor — its group, type, URL, latest status, last ping, 24-hour uptime and page config — ready for a spreadsheet, a database, an uptime/SLA report or an observability pipeline.

Built for agents

• A personal operations agent can check whether home-lab or business services are up before starting a dependent workflow.
• An incident agent can turn public status pages into structured monitor state, latency and 24-hour uptime for triage or alerts.
• A research agent can compare public Uptime-Kuma fleets without scraping dashboards or holding credentials.

The ready-to-run example targets the official Uptime-Kuma demo and returns five live monitors. The form caps the first run at 20 rows, so it is safe to call from an autonomous tool loop.

Uptime-Kuma only.

What it does

• Any Uptime-Kuma page, one input. Pass the domain base (https://status.kuma.pet). The status-page slug is resolved automatically via GET /api/entry-page, or read from a …/status/<slug> UI URL, or set explicitly with slug.
• Multiple sources per run. Give a list of status pages; each is harvested independently.
• Three modes.
  • All (default) — every monitor from the config tree, with latest status / ping / 24h uptime joined from the heartbeat API where available.
  • Heartbeat — only monitors that actually have a heartbeat series (status/ping/uptime populated).
  • Monitors — the monitor tree only; no heartbeat request is made (fastest, config-only).
• Reads by key presence. Uptime-Kuma instances differ only in which optional fields they populate. A port / dns / ping monitor with no url, a page with no open incident, a version that stores incidents under incident (singular) vs incidents (plural) — every one yields null for that column, never an error.
• Harvests the open pages. A private or unpublished page 404s its slug — that source is skipped with a logged note and the run continues, never crashing.

Input

Field	Required	Description
sources	✅	A list of Uptime-Kuma status-page base URLs (e.g. ["https://status.kuma.pet"]). A …/status/<slug> UI URL also works — the slug is read from the path.
slug		The status-page slug to use for every source. Leave empty to resolve per source via /api/entry-page. Set it when a host serves multiple pages.
mode		all (default), heartbeat, or monitors. See modes above.
maxRecords		Global cap on emitted monitor rows across all sources (each row = one billable event).

Example inputs

{ "sources": ["https://status.kuma.pet"] }

{ "sources": ["https://status.kuma.pet"], "slug": "uptime-kuma", "mode": "all" }

{ "sources": ["https://status.nixnet.services"], "mode": "monitors" }

{ "sources": ["https://status.kuma.pet", "https://status.nixnet.services"], "mode": "heartbeat", "maxRecords": 500 }

Output

One row per monitor:

Column	Description
source	The status page origin the monitor came from.
slug	The resolved status-page slug.
groupId / groupName	The parent group the monitor belongs to (e.g. Web services).
monitorId	Uptime-Kuma's numeric id for the monitor.
name	The monitor's display name.
type	Monitor type (http, keyword, port, dns, ping, …).
url	The monitored URL, for monitors that expose one (null for port/dns/… checks).
sendUrl	Whether the page shows the monitor's URL publicly (1/0).
pageTitle / theme / published	Page-level config, promoted onto every row.
openIncidentTitle	Title of the first open incident on the page, else null.
lastStatus	Latest heartbeat status: 1 = up, 0 = down, 2 = pending, 3 = maintenance. null in monitors mode.
lastPing	Latest heartbeat response time in milliseconds (null when the check records none).
lastHeartbeatTime	Timestamp of the latest heartbeat.
uptime24h	24-hour uptime ratio (1 = 100%, 0.9874 = 98.74%).
heartbeatCount	How many heartbeats the page returned for the monitor.
_monitorCount	How many monitors the source page publishes.
_raw	The complete original monitor object — nothing is ever dropped.

Every field is nullable and read by key presence — an instance that omits url / an incident / a heartbeat simply yields null for that column, never an error.

Notes & scope

• Public status data only. This reads a status page's public JSON — the same data the dashboard renders in a browser. No login is used or required.
• Honest failure. A page with no monitors, or every source private, returns 0 records and a clean success. A URL that isn't an Uptime-Kuma status page (an HTML page, a body with no publicGroupList) fails fast with a clear message.
• Find instances. Any public Uptime-Kuma status page works; the demo at https://status.kuma.pet is a good place to try it.

Pricing

Pay-per-event: $0.0005 per monitor returned. The current five-monitor demo is $0.0025 at list price; the 20-row first-run cap is at most $0.01. A capped run charges only for what it emits.